WO2024205445A1 - Method for ai-based situational analysis in education - Google Patents

Abstract

The invention relates to the field of computing. A hardware and software system for mapping a personalized learning path for a user comprises the following interlinked elements: a system for collecting external information about a user, which is configured to collect information from publicly available sources of information according to key parameters; a system for collecting internal information about the user, which is configured to collect information on the basis of additional data provided during user registration and completion of a questionnaire; a system for mining the data obtained from the aforementioned systems, which is configured to analyze the data associated with a user in order to determine their level of competence in different disciplines; a system for generating recommendations, which is configured to generate a list of recommendations on the basis of an analysis of the retrieved data; and a system for storing data, said system being a highly secure converged database hosted on a cloud server platform.

Description

CASE STUDY ANALYSIS METHOD WITH AI IN TRAINING

AREA OF TECHNOLOGY

This technical solution relates to the field of computer technology, in particular, to a software and hardware complex for constructing an individual user learning trajectory.

LEVEL OF TECHNOLOGY

The prior art discloses a solution selected as the closest analogue, RLI 2672171 C1. This solution is a method for processing information to prepare recommendations based on a computerized assessment of the user's abilities, in which, using authentication tools that provide access to the user's device to solve at least one computerized task, the user's emotions are recognized in real time.

The above analogue characterizes the method of processing information for preparing recommendations based on a computerized assessment of user abilities, in which: access is provided using authentication means to the solution of at least one computerized task to the user's device; recognition of the user's emotions is carried out in real time during said solution using one video camera, with the help of which a set of emotions is obtained and artificial neural networks are used to recognize the received emotions; data is transmitted to the data processing device of the system for obtaining the parameters of the user's actions during said solution of at least one task; at least one user ability is determined based on the data obtained after the user's solution for one task and the recognized user emotions; in the data processing device, the data is converted into numerical values, and the recognized emotions are converted into a scale, according to which the user's ability indicator is calculated; based on the results obtained, the ability determination system generates recommendations for the user to make decisions based on at least one ability determined in the previous step based on the calculated ability values.

The proposed technical solution is aimed at eliminating the shortcomings of the current level of technology and differs from known solutions in that the proposed software and hardware complex ensures high-quality and effective construction of an individual user learning trajectory. ESSENCE OF THE INVENTION

The technical problem that the claimed solution is aimed at solving is the creation of a software and hardware complex for constructing an individual user learning trajectory. Additional embodiments of the present invention are presented in the dependent claims of the invention.

The technical result consists in improving the quality of user training due to the effective construction of an individual user training trajectory.

The stated result is achieved by implementing a hardware and software complex for constructing an individual user training trajectory, which contains the following interconnected components: a system for collecting external information about the user, designed with the ability to collect information on key parameters from publicly available sources of information; a system for collecting internal information about the user, designed with the ability to collect information based on additional data specified during registration and survey of the user; a system for intelligent analysis of the data received from the above-mentioned systems, designed with the ability to analyze data associated with the user, determining the level of training in areas; a system for generating recommendations, designed with the ability to generate a list of recommendations based on the analysis of the extracted data; a data storage system, wherein the system is a convergent database with a high degree of protection, which is deployed on a cloud server platform.

In a particular version of the implementation of the described complex, the recommendation generation system uses a method of filtering restrictions, which outputs recommendations only for the most complete data, cutting off the least filled categories.

In a particular version of the implementation of the described complex, in the system of intelligent analysis of the obtained data, the directions are: grammar; mathematical knowledge; general educational training; communication; stress level; interests; strengths; weaknesses; goals.

DETAILED DESCRIPTION OF THE INVENTION

In the following detailed description of the embodiment of the invention, numerous implementation details are set forth in order to provide a clear understanding of the present invention. However, it will be apparent to one skilled in the art how the present invention may be used with or without these implementation details. In other instances, well-known methods, procedures, and components have not been described in detail in order not to unnecessarily obscure the features of the present invention.

In addition, it will be clear from the above description that the invention is not limited to the embodiment shown. Numerous possible modifications, changes, variations and substitutions, preserving the essence and form of the present invention, will be obvious to those skilled in the art.

Additional decision support systems and individual learning path building play an important role in the modern educational environment. Initially, educational service providers concentrated their efforts on creating a large number of educational products in order to generate income and expand their customer base.

However, it is clear that users have begun to mistrust the information around them, as the market has almost reached the saturation point. As a result, educational service providers have switched to providing users with services for individual selection of learning paths to achieve the user's goals in order to increase their income.

This solution is designed to solve the problems of situational analysis in training, in which management decisions are made at each stage depending on the results of the previous stage. The innovativeness of the solution lies in the processing of internal and external information when working with situations, to obtain the most accurate analysis result. Internal information is obtained from decisions made within the case, external information includes the processing of multimedia information, spelling information, taking into account the gender, age, education and social status of the student, statistical processing of social activity.

The proposed technical solution is used in secondary and higher education organizations, as well as in the commercial sector. The solution allows performing cognitive tasks, helps to build an individual learning path to achieve user goals.

The present technical solution, namely the software and hardware complex for constructing an individual user learning trajectory, contains the following interconnected technical elements: a system for collecting external information about the user, designed with the ability to collect information on key parameters from publicly available sources of information; a system for collecting internal information about the user, designed with the ability to collect information based on additional data specified during registration and user survey; a system for intelligent analysis of the data received from the above-mentioned systems; a system for generating recommendations; a data storage system.

The external information collection system is capable of collecting data from the user's social networks, transferring the obtained data to a data warehouse, from which the data is transferred to the intelligent analysis system for further processing. The system, in the preferred embodiment, is implemented as a database server. The external information collection system selects the collected data by words, terms, phrases, quotes or assessments that may be related to the required classifiers of the collected data about the user. The data is structured by classification cells. Structuring occurs according to the following thirteen parameters: completeness of personal information; education; interests; work experience; knowledge of foreign languages; frequency of posting information; family ties; social activity; involvement of third-party users in the user's social activity; ratio of posted text information to graphic information; grammar; user's emotional mood based on the frequency of use of tagged words in their posts; use of obscene language. Parameters are divided into subcategories based on the depth at which the system can manipulate the data selection based on configurable policies. For example, a configurable policy can specify not only the user's level of education, but also the number of educational institutions, the rating of institutions. A configurable policy sets the depth of data selection and the removal of unnecessary data after structuring.

The structured data is passed on for further analysis and storage.

The internal information collection system performs the function of classifying the user based on the data provided by him. The system, in the preferred embodiment, is implemented as a database server.

The internal information collection system divides the received data into attributes, and divides the attributes by features, after which it distributes the received information by the degree of similarity before assigning it to one of the pre-known classes with the parameters available in the classifier.

When registering, the user enters the following information about themselves into the database: gender; age; level of education; availability of additional education; frequency of studying new educational programs. After receiving the data, the user is asked to complete a survey containing questions to identify general aspects of the user's level of training, their interests, current level of satisfaction with their knowledge, goals for the level of knowledge (the data is recorded in the database). The volume of the survey depends on the algorithm achieving a "clean" classification of the answer's relation to one of the categories. Additional corrective questions with multiple choice options are allowed. The algorithm for reproducing corrective questions reproduces questions until a "clean" response ratio is obtained in one of the specified subcategories. The duration of the survey is from three minutes. To reduce the time for completing the survey, the maximum permissible number of corrective questions for each category is specified in the algorithm for reproducing corrective questions. The "clean" ratio upon reaching the maximum permissible number of corrective questions is based on the numerical number of responses related to one of the categories. During the survey, graphic elements appear to support the user's emotional involvement in the process.

Intelligent analysis system for received data.

The data collected from the above mentioned sources are combined and consolidated for entry into the data mining system, which is a set of data on the state of the user being studied, applies a systematic comprehensive approach to studying the results of the user's activity, summarizes the analysis materials in the form of conclusions and recommendations based on the processing of all available information on the result of the collected data.

The system uses connectors in the form of API interfaces, protocols, and services to communicate with data sources and the data storage system. The system, in the preferred embodiment, is implemented on computing server hardware and uses its resources to implement the specified tasks.

The data mining algorithm uses the decision tree machine learning method, which allows for high-quality and resource-intensive analysis of external sources and kNN (nearest neighbor method) when working with internal data sources.

The algorithm performs cognitive tasks: it receives data, combines input and output simultaneously at each step of the user's action to "learn" the data, and based on the data set and the desired result, produces output when new input data is entered.

The construction proceeds from the root node to the leaves. Gradually dividing the received data based on certain attributes into subgroups, the closest relation to a certain classification is achieved. At least one command makes the system receive a request for a recommendation. At least one command makes it analyze the data associated with the user, determining the level of preparation in the areas of: grammar; mathematical knowledge; general education; communication; stress level; interests; strengths; weaknesses; goals. The recommendation generation system compares the characteristics of the data obtained from the intelligent analysis system to the most attractive options for increasing the ratio of data for each parameter to the full value. The system, in the preferred embodiment, is implemented on computing server equipment and uses its resources to implement the established tasks.

At least one team is forced to create a list of recommendations based on the analysis of the extracted data. The set of recommendations uses a filtering constraint method that outputs recommendations only for the most complete data, cutting off the least complete categories. The higher the volume of extracted data, the more successful the task will be and the higher the predictive accuracy in constructing an individual learning path.

Based on the received parameters, the system builds a chain of personalized tasks, which consist of instructional information materials displayed in the user interface in the form of training elements. Training elements are electronic information carriers (for example, plans for thematic lessons, texts, videos, graphic elements) filled with information to improve the level of training in each area.

The resulting data is stored in a distributed data storage system where it can be stored for long periods of time and duplicated to ensure reliability.

The system is a highly secure converged database deployed on the Oracle cloud server platform. The system volume is optimized to increase the speed of operation. The present technical solution, in a preferred embodiment, is implemented by means of at least databases, computing equipment (for example, a server (cloud or physical)), display devices (for example, displays, tablets), as well as computing devices by means of which the student interacts through a user interface.

In this way, a high-quality and effective construction of an individual learning trajectory for the user is realized.

In these application materials, a preferred disclosure of the implementation of the claimed technical solution was presented, which should not be used as limiting other, particular embodiments of its implementation that do not go beyond the requested scope of legal protection and are obvious to specialists in the relevant field of technology.

Claims

Formula 1. A hardware and software complex for constructing an individual user learning trajectory, comprising the following interconnected systems: a system for collecting external information about the user, capable of collecting information on key parameters from publicly available sources of information; a system for collecting internal information about the user, capable of collecting information based on additional data specified during registration and user survey; a system for intelligent analysis of the data received from the above-mentioned systems, capable of analyzing data associated with the user, determining the level of training in areas; a system for generating recommendations, capable of generating a list of recommendations based on the analysis of the extracted data; a data storage system, wherein the system is a convergent database with a high degree of protection, which is deployed on a cloud server platform. 2. The hardware and software complex according to item 1, in which the recommendation generation system uses a filtering constraint method that outputs recommendations based on the most complete data, cutting off the least filled categories. 3. The hardware and software complex according to item 1, wherein in the system of intellectual analysis of the received data, the directions are: grammar; mathematical knowledge; general educational preparation; communication; stress level; interests; strengths; weaknesses; goals. 9